Skip to content

image_processing

Directory actions

More options

Directory actions

More options

Latest commit

 

History

History

image_processing

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
img
img
 
 
Image Processing Dashboard.json
Image Processing Dashboard.json
 
 
README.md
README.md
 
 

README.md

Image Processing Pipeline

Architecture

robustness

The system is composed by a manager node and a number of format, resolution and size worker nodes.

The work is distributed through messaging, but the files are accessed through a shared file system.

Specifications

Processes

  • Format: Converts an image into png format.
  • Resolution: Resizes an image to a given resolution. The chosen resolution is: 100x100
  • Size: Resizes an image to a given size. The chosen size is: 30x30

Data

BIRDS 525 SPECIES dataset will be used for the benchmarks. This dataset is further augmented using image_multiplicator.py

Benchmarks

Benchmarks will be run using 2, 4 and 6 worker nodes (of each type).

  • Measurements
    • Completion Time
    • CPU usage
    • Node throughput and Combined throughput
    • Memory usage
    • Network usage
    • Coefficient of variation of worktime and results produced

The metrics dashboard can be found here